agent-context

Bootstrap persistent project context for AI coding agents across sessions.

23|3|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/AndreaGriffiths11/agent-context-system --skill agent-context
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-context
Source: https://github.com/AndreaGriffiths11/agent-context-system/tree/main/openclaw
Command: npx skills add https://github.com/AndreaGriffiths11/agent-context-system --skill agent-context

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill preserves long-running project knowledge between sessions by maintaining a persistent project context (AGENTS.md) and a private scratchpad, ensuring that learning from one session influences subsequent work.

Core Features & Use Cases

  • Persistent context across sessions: keeps project knowledge in AGENTS.md and a personal .agents.local.md scratchpad for continuous learning.
  • Automated lifecycle workflow: supports init, compression, and promotion steps to maintain concise, relevant context.
  • Offline, self-contained data: stores all context in Markdown files shipped with the repository, with no external dependencies.
  • Use case: a developer begins work on a new project and the agent recalls conventions, patterns, and decisions from prior sessions to accelerate setup and coding.

Quick Start

Initialize in your project by running agent-context init

Frequently Asked Questions about agent-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I preserve project context for AI coding agents across sessions?▼

You can preserve project context across coding sessions by maintaining a persistent AGENTS.md knowledge base and a private .agents.local.md scratchpad, ensuring that learning from one session continuously influences subsequent work.

What is the best way to bootstrap persistent context for software projects?▼

The best way to bootstrap persistent context is running the agent-context init command, which sets up a Markdown knowledge base to store patterns, conventions, and decisions for rapid agent onboarding and safer automation.

Does this persistent context tool require external dependencies or databases?▼

No external dependencies or databases are required. The persistent context tool stores all project knowledge offline in self-contained Markdown files that are shipped directly within the repository.

How do I manage context lifecycle and compression for my coding agents?▼

You manage context lifecycle through automated init, compression, and promotion steps that maintain concise, relevant Markdown context, preventing knowledge files from becoming bloated over multiple coding sessions.

Can I use Markdown files to store offline knowledge for my coding agents?▼

Yes, you can use Markdown files to store offline knowledge. The tool loads Markdown as context and supports optional directories for scripts, references, and assets directly within your repository.

When do I need persistent project context for automated coding workflows?▼

You need persistent project context when running automated coding workflows where agents must remember patterns, conventions, and decisions between runs, enabling rapid onboarding and safer automation for new software projects.